A Fuji Electric FRN90AQ1L-4E drive caused repeated residual-current trips on an NSX400X fitted with an adjustable Vigi unit. The evidence distinguishes expected drive leakage at very sensitive settings from the abnormal condition ultimately confirmed by replacing the drive.
Observed NSX400X Trip Pattern
The installation experienced approximately one trip per week for three to four weeks, always between 2:00 a.m. and 4:00 a.m. The operator progressively increased the trip delay to its stated maximum setting of 320 ms, while the Vigi sensitivity was set to 1 A.
| Test or setting | Observed result |
|---|---|
| Drive power disconnected; Vigi at 30 mA | No trip |
| Drive power connected; Vigi at 30 mA | Immediate trip, even without a run demand |
| Original drive; Vigi at 1 A and 320 ms | The booster operated during the day, but sensitivity adjustments could still provoke a trip |
| Replacement FRN90AQ1L-4E; Vigi at 500 mA | No further trips were reported |
Separate Expected Leakage from a Drive Fault
A powered variable-frequency drive can produce earth-leakage current without running the motor. Possible paths identified in the evidence include the input EMC filter and capacitive currents associated with the drive, motor cable, or motor under high output-voltage slew rates. Cable length can influence these currents. Therefore, operation at 30 mA alone did not prove that the drive had failed.
A reported instruction-manual value called for a Vigi setting of at least 440 mA for this application. Treat that value as a threshold requiring confirmation against the applicable manufacturer documentation and installed protection type. A trip above that threshold, especially when comparison with an identical drive removes the fault, supports a defective-drive diagnosis.
Diagnostic Procedure
- Record the Vigi sensitivity and delay before changing them. Do not use a longer delay as the sole remedy for recurring unexplained trips.
- Confirm which powered branch causes the trip. In this case, the system remained stable at 30 mA with drive power removed and tripped when drive power was restored.
- Check the residual-current protection type and verify its suitability for a variable-frequency drive. Also assess the input EMC filter, motor-cable length, cable condition, and motor insulation as possible leakage contributors.
- Interpret the load insulation test separately. The booster insulation test found nothing abnormal, but that result did not exclude leakage or an internal fault on the drive input side.
- If the drive is self-powered and cannot be subdivided safely for internal isolation tests, substitute a known serviceable drive of the same model where site procedures permit. Repeat the test without changing unrelated wiring.
Confirmed Resolution and Verification
The original 400 V, 90 kW FRN90AQ1L-4E was replaced with a used spare of the same model. With the Vigi set to 500 mA, the installation stopped tripping. Because the wiring and application remained in service while replacement removed the symptom, the final field diagnosis was a failed original drive.
Verify the repair by confirming stable energization with no run demand, normal booster operation, and absence of the previous overnight trips. Retain the final 500 mA setting only after checking the required personnel and fire-protection objectives and the applicable manufacturer documentation.
FAQ
Why does the NSX400X trip at 30 mA when the Fuji drive is not running?
The drive is still energized and can pass earth-leakage current through its internal EMC filtering even without a run demand. In this case, removing drive power stopped the 30 mA trip.
What Vigi sensitivity was reported for the FRN90AQ1L-4E?
The cited instruction-manual threshold was at least 440 mA, and the resolved installation used 500 mA. Confirm the required setting and protection type against the applicable manufacturer documentation before adoption.
How was the faulty Fuji FRN90AQ1L-4E confirmed?
The original drive was replaced with another unit of the same model. With the Vigi set to 500 mA, no further trips were reported, while the booster insulation test had shown no particular defect.
